A Rolling Bearing Remaining Life Prediction Method Based on Wiener Process Data-Model Integration in the Context of Mechanical Fault Diagnosis

Data-driven methods for predicting the remaining service life of rolling bearings rely too much on training data and lack the necessary theoretical foundation, resulting in poor interpretability. To address this problem, this paper proposes a model-data linkage method based on wiener process, which uses the generalized wiener process to describe the degradation process of the bearing health state and utilizes a particle filtering algorithm to dynamically match the collected data to predict the remaining service life of rolling bearings. The proposed method is validated in the accelerated life experimental dataset of XJTU-SY rolling bearings, and compared with the separate wiener process-based and data-driven association vector machine methods and the model-based particle filter. The average root mean square error of the proposed method is 13.74, and the average absolute error is 9.64, which is much lower than that of the other three methods, and it confirms the effectiveness of the method in this paper.